U.S. patent application number 13/003653 was filed with the patent office on 2011-05-05 for control method of wireless communication system, wireless communication system, transmitting apparatus, and receiving apparatus.
This patent application is currently assigned to NEC CORPORATION. Invention is credited to Kenichi Hosoya, Kenichi Maruhashi, Naoyuki Orihashi.
Application Number | 20110105032 13/003653 |
Document ID | / |
Family ID | 41550129 |
Filed Date | 2011-05-05 |
United States Patent
Application |
20110105032 |
Kind Code |
A1 |
Maruhashi; Kenichi ; et
al. |
May 5, 2011 |
CONTROL METHOD OF WIRELESS COMMUNICATION SYSTEM, WIRELESS
COMMUNICATION SYSTEM, TRANSMITTING APPARATUS, AND RECEIVING
APPARATUS
Abstract
A channel response matrix is obtained by performing a training
process between a transmitter 401 and a receiver 402 to obtain
optimal signal phases of the antenna array. Next, a singular-value
decomposition (SVD) process is performed to decompose the channel
response matrix into a correlation matrix and eigenvalues. Next, a
diagonal matrix having square roots of the eigenvalues as its
components is obtained. Next, all but one of diagonal components
included in the diagonal matrix are replaced with zeros, and
optimal setting of the amplitudes and phases of signals to be
applied to the antenna array (antenna weight vector) for use in
wireless communication between the transmitter and the receiver is
obtained based on a channel response matrix that is reconstructed
by using the component-replaced diagonal matrix. In this way, when
wireless communication is implemented by performing beam forming,
the time necessary to find and set a beam direction can be
reduced.
Inventors: |
Maruhashi; Kenichi; (Tokyo,
JP) ; Orihashi; Naoyuki; (Tokyo, JP) ; Hosoya;
Kenichi; (Tokyo, JP) |
Assignee: |
NEC CORPORATION
|
Family ID: |
41550129 |
Appl. No.: |
13/003653 |
Filed: |
May 13, 2009 |
PCT Filed: |
May 13, 2009 |
PCT NO: |
PCT/JP2009/002089 |
371 Date: |
January 11, 2011 |
Current U.S.
Class: |
455/59 |
Current CPC
Class: |
H04B 7/0617
20130101 |
Class at
Publication: |
455/59 |
International
Class: |
H04B 7/04 20060101
H04B007/04 |
Foreign Application Data
Date |
Code |
Application Number |
Jul 16, 2008 |
JP |
2008-184405 |
Claims
1. A method of controlling a wireless communication system which
comprises a transmitter comprising a transmitting antenna array and
a receiver comprising a receiving antenna array, wherein when
amplitudes and phases of signals to be transmitted from at least
two antenna elements among a plurality of antenna elements
constituting the transmitting antenna array are independently
controlled and amplitudes and phases of signals to be received at
least two antenna elements among a plurality of antenna elements
constituting the receiving antenna array are independently
controlled, the method comprises: obtaining a channel response
matrix by performing a training process to obtain an optimal
setting of amplitudes and phases of signals to be applied to the
antenna array (hereinafter called "antenna weight vector") at least
between the transmitter and the receiver; performing singular-value
decomposition process to decompose the channel response matrix into
a correlation matrix and eigenvalues; obtaining a diagonal matrix
having square roots of the eigenvalues obtained in the
singular-value decomposition process as its components; and
replacing all but one of diagonal components included in the
diagonal matrix with zeros, and obtaining an antenna weight vector
to be applied to the antenna array for use in wireless
communication between the transmitter and the receiver based on
channel response matrix reconstructed by using the
component-replaced diagonal matrix.
2. The method of controlling a wireless communication system
according to claim 1, wherein when the antenna weight vector to be
applied to the antennal array is obtained, the method further
comprises: obtaining a plurality of reconstructed channel response
matrixes; and individually obtaining an antenna weight vector to be
applied to the antenna array for each of the plurality of
reconstructed channel response matrixes based on the reconstructed
channel response matrixes, and using one of the plurality of
individually-obtained antenna weight vectors as the optimal antenna
weight vector to be applied to the antenna array.
3. The method of controlling a wireless communication system
according to claim 2, further comprising assigning priority ranks
to the antenna weight vectors, which were individually obtained for
each of the plurality of reconstructed channel response matrixes,
in descending order of magnitude of the diagonal components,
selecting an antenna weight vector to be applied to the antenna
array of the transmitter and the antenna array of the receiver
according to the priority ranks, and performing wireless
communication by using the selected antenna weight vector.
4. The method of controlling a wireless communication system
according to claim 3, further comprising selecting an antenna
weight vector according to the priority ranks in response to
deterioration in the communication quality between the transmitter
and the receiver, and performing wireless communication by applying
the selected antenna weight vector.
5. The method of controlling a wireless communication system
according to claim 1, wherein a first radio wave used to carry a
training signal sent from the transmitter to the receiver for the
training process has a narrower transmission frequency band than
that of a second radio wave used for transmission of an information
signal that is performed by applying the optimal antenna weight
vector to be applied to the antenna array, or the first radio wave
is modulated by a modulation method having a larger distance
between signal points than that of the second radio wave.
6. A wireless communication system comprises: a transmitter
comprising a transmitting antenna array comprising a plurality of
transmitting antennal components, the transmitter being configured
to independently control amplitudes and phases of transmission
signals to be transmitted from at least two antenna elements among
the plurality of antenna elements; and a receiver comprising a
receiving antenna array comprising a plurality of receiving
antennal components, the receiver being configured to independently
control amplitudes and phases of received signals to be received at
least two antenna elements among the plurality of antenna elements
constituting the receiving antenna array, wherein the transmitter
and the receiver are configured so as to perform an amplitude/phase
control process of the transmitting and receiving antenna arrays in
cooperation the amplitude/phase control process comprises:
obtaining a channel response matrix by performing a training to
obtain an optimal antenna weight vector to be applied to the
antenna array at least between the transmitter and the receiver;
performing singular value decomposition to decompose the channel
response matrix into a correlation matrix and eigenvalues;
obtaining a diagonal matrix having square roots of the eigenvalues
obtained by the singular-value decomposition as its components;
obtaining an antenna weight vector to be applied to the antenna
array for use in wireless communication between the transmitter and
the receiver based on a channel response matrix reconstructed by
using a component-replaced diagonal matrix, the component-replaced
diagonal matrix being obtained by replacing all but one of diagonal
components included in the diagonal matrix with zeros; and
controlling amplitudes and phases of the transmission signals and
received signals in accordance with the antenna weight vector.
7. The wireless communication system according to claim 6, wherein
said obtaining the antenna weight vector to be applied to the
antennal array comprises: obtaining a plurality of reconstructed
channel response matrixes; obtaining individually an antenna weight
vector to the applied to the antenna array for each of these
plurality of reconstructed channel response matrixes based on the
reconstructed channel response matrixes; and using one of the
plurality of individually-obtained antenna weight vectors as the
optimal antenna weight vector to be applied to the antenna
array.
8. The wireless communication system according to claim 7, wherein
the amplitude/phase control process further comprises: storing the
plurality of antenna weight vectors, which were individually
obtained for each of the plurality of reconstructed channel
response matrixes, as a data string; obtaining a new channel
response matrix by performing the training at least either at
predetermined intervals or at random timing during communication
between the transmitter and the receiver; and updating the stored
data string according to timing of a calculation of the new channel
response matrix.
9. The wireless communication system according to claim 7, wherein
the amplitude/phase control process further comprises: assigning
priority ranks to the antenna weight vectors, which were
individually obtained for each of the plurality of reconstructed
channel response matrixes, in descending order of magnitude of the
diagonal components; selecting an antenna weight vector to be
applied to the transmitting antenna array and the receiving antenna
array according to the priority ranks; and performing wireless
communication by using the selected antenna weight vector.
10. The wireless communication system according to claim 9, wherein
said selectin the antenna weight vector in response to
deterioration in communication quality between the transmitter and
the receiver.
11. The wireless communication system according to claim 6, wherein
a first radio wave used to carry a training signal sent from the
transmitter to the receiver for the training process has a narrower
transmission frequency band than that of a second radio wave used
for transmission of an information signal that is performed by
applying the optimal antenna weight vector to be applied to the
antenna array, or the first radio wave is modulated by a modulation
method having a larger distance between signal points than that of
the second radio wave.
12. The wireless communication system according to claim 6, wherein
a radio wave having a frequency equal to or higher than 10 GHz is
used for the wireless communication.
13. A transmitting apparatus that performs communication with a
receiving apparatus, comprising: a transmitting antenna array
comprising a plurality of antenna elements; and control unit
adapted to change changing a beam direction of the transmitting
antenna array by controlling amplitudes and phases of signals to be
transmitted from at least two antenna elements among the plurality
of antenna elements, wherein the control unit adjusts the beam
direction by supplying one antenna weight vector selected from a
plurality of antenna weight vectors to the transmitting antenna
array, and performs control such that the antenna weight vector to
be supplied to the transmitting antenna array is switched to a
different one of the plurality of antenna weight vectors in
response to deterioration in communication quality with the
receiving apparatus, and each of the plurality of antenna weight
vectors corresponds to one of a plurality of eigenpaths of a radio
transmission path between the transmitting apparatus and the
receiving apparatus, the plurality of eigenpaths being obtained by
performing a singular-value decomposition of a channel response
matrix with regard to the radio transmission path.
14. A receiving apparatus that performs communication with a
transmitting apparatus, comprising: a receiving antenna array
comprising a plurality of antenna elements; and control unit
adapted to change a beam direction of the receiving antenna array
by controlling an antenna weight vector of a signal to be received
by at least two antenna elements among the plurality of antenna
elements, wherein the control unit adjusts the beam direction by
supplying one antenna weight vector selected from a plurality of
antenna weight vectors to the receiving antenna array, and performs
control such that the antenna weight vector to be supplied to the
receiving antenna array is switched to a different one of the
plurality of antenna weight vectors in response to deterioration in
communication quality between the transmitting apparatus, and each
of the plurality of antenna weight vectors corresponds to one of a
plurality of eigenpaths of a radio transmission path between the
transmitting apparatus and the receiving apparatus, the plurality
of eigenpaths being obtained by performing a singular-value
decomposition of a channel response matrix with regard to the radio
transmission path.
15. A wireless communication system comprises: a transmitter
comprising a transmitting antenna array comprising a plurality of
transmitting antennal components, the transmitter being configured
to independently control amplitudes and phases of transmission
signals to be transmitted from at least two antenna elements among
the plurality of antenna elements; a receiver comprising a
receiving antenna array comprising a plurality of receiving
antennal components, the receiver being configured to independently
control amplitudes and phases of received signals to be received at
least two antenna elements among the plurality of antenna elements
constituting the receiving antenna array, channel response matrix
calculation means for obtaining a channel response matrix by
performing a training to obtain an optimal antenna weight vector to
be applied to the antenna array at least between the transmitter
and the receiver; singular-value decomposition means for
decomposing the channel response matrix into a correlation matrix
and eigenvalues; and amplitude/phase control means for obtaining a
diagonal matrix having square roots of the eigenvalues obtained by
the singular-value decomposition means as its components, replacing
all but one of diagonal components included in the diagonal matrix
with zeros, obtaining an antenna weight vector to bc applied to the
antenna array for use in wireless communication between the
transmitter and the receiver based on a channel response matrix
reconstructed by using the component-replaced diagonal matrix, and
controlling amplitudes and phases of the transmission signals and
received signals in accordance with its result.
Description
TECHNICAL FIELD
[0001] The present invention relates to a technical field of
wireless communication in which wireless communication is
implemented by using radio beams that are set based on the
communication quality.
BACKGROUND ART
[0002] In recent years, wireless devices using wide-band millimeter
waves (30 GHz to 300 GHz) have become widespread. The
millimeter-wave radio technology has been expected to be especially
applicable to Gigabit-class high-rate radio data communication such
as radio transmission of high-resolution images (for example, see
Non-patent literatures 1 and 2).
[0003] However, the millimeter waves, which have high frequencies,
have a high rectilinear propagation property, and thus raising a
problem when radio transmission is to be implemented indoors. In
addition to having the high rectilinear propagation property, the
millimeter-wave signals are significantly attenuated by a human
body or other objects. Therefore, when a person stands between the
transmitter and the receiver in a room or the like, it is
impossible to obtain an unobstructed view, thus making the
transmission very difficult (shadowing problem). Since this problem
is caused as a result of the higher rectilinear propagation
property of radio waves resulting from the use of higher
frequencies as well as the change in the propagation environments,
the problem is not limited to the millimeter wave band (30 GHz and
above). Although it is not easy to clearly specify the transition
frequency, it has been said to be around 10 GHz. Meanwhile,
according to recommendations of the International
Telecommunications Union ("Propagation data and prediction methods
for the planning of indoor radio communication systems and radio
local area networks in the frequency range 900 MHz to 100 GHz,"
ITU-R, P.1238-3, April, 2003), the power loss coefficient, which
indicates the attenuation amount of a radio wave with respect to
the propagation distance, is 22 for 60 GHz in an office, while it
is 28 to 32 for 0.9 to 5.2 GHz. Considering that it is 20 in the
case of free-space loss, the effects of scattering, diffraction,
and the like are considered to be small for high frequencies in the
order of 60 GHz.
[0004] To solve the problem described above, for example, Patent
literature 1 discloses a system in which more than one transmission
path is provided by installing a plurality of receiving units in
the receiving device, so that when one of the transmission paths
between the transmitting device and the receiving units is blocked,
the transmission is performed by another transmission path.
Furthermore, as another method for solving the problem, Patent
literature 2 discloses an invention to secure plural transmission
paths by installing reflectors on walls and a ceiling.
[0005] In the method disclosed in Patent literature 1, it is very
difficult to continue the communication when the area at and around
the transmitting device is shielded or when all of the installed
receiving units are shielded. Meanwhile, the method disclosed in
Patent literature 2 requires the user to take the trouble to
install the reflectors with consideration given to the positions of
the transmitter and the receiver and the like.
[0006] However, recent studies on the propagation properties of
millimeter waves have found out that there is a possibility that
reflected waves can be utilized without intentionally installing
the reflectors. FIG. 9 is a schematic diagram of a communication
system using a millimeter wave band. Each of a transmitter 91 and a
receiver 92 has a wide-angle antenna. FIG. 10 shows an example of a
delay profile of the system using the wide-angle antennas shown in
FIG. 9 when the system is used indoors. In the system using the
wide-angle antennas shown in FIG. 9, the received power of the
dominant wave, which is arrives faster than any other waves, is
larger than that of any other waves as shown in FIG. 10. After
that, although delayed waves such as the second and third waves
arrive, the received power of these waves is smaller than that of
the dominant wave. These second and third waves are reflected waves
from the ceiling and the walls. This situation is remarkably
different from the propagation environment of radio waves having a
lower rectilinear propagation property, such as 2.4 GHz band used
in wireless LANs (Local Area Networks). In 2.4 GHz band, it is very
difficult to clearly separate waves in their directions of Arrival
because of the effects of diffraction and multiple reflections. In
contrast to this, in the millimeter waves having a high rectilinear
propagation property, although radio waves are relatively clearly
distinguished in their directions of Arrival, the number of delayed
waves is limited and the received-signal level of the delayed waves
is relatively small.
[0007] Therefore, in communication systems using a frequency band
around or higher than 10 GHz such as millimeter waves, when the
direct wave (dominant wave) is shielded, the receiver must point a
narrow beam having a high directive gain to the direction of
Arrival of a reflected wave to ensure a sufficient received-signal
level so that the transmission can be continued by using the
reflected wave. However, in order to eliminate the necessity for
the user to take the trouble in regard to the relative positions of
the transmitter and receiver, and the like, the beam forming
technology capable of dynamically controlling the direction of a
narrow beam is indispensable.
[0008] In the beam forming, it is necessary to construct an antenna
array. For millimeter waves having a short wavelength (e.g., 5 mm
in the case of frequency of 60 GHz), the antenna array can be
implemented in a small area. Phase shifter arrays and oscillator
arrays for use in such antenna arrays for millimeter waves have
been developed (for example, see Non-patent literatures 3 and
4).
Citation List
Patent Literature
[0009] Patent Literature 1: Japanese Unexamined Patent Application
Publication No. 2006-245986 [0010] Patent Literature 2: Japanese
Unexamined Patent Application Publication No. 2000-165959 [0011]
Patent Literature 3: US Patent Publication No. 2007/0205943
Non Patent Literature
[0011] [0012] Non Patent Literature 1: K. Maruhashi et al.,
"60-GHz-band LTCC Module Technology for Wireless Gigabit
Transceiver Applications", IEEE International Workshop on
Radio-Frequency Integration Technology, Digest, pp. 131-134,
December 2005. [0013] Non Patent Literature 2: K. Ohata et al.,
"1.25 Gbps Wireless Gigabit Ethernet Link at 60 GHz-Band", IEEE
MTT-S International Microwave Symposium, Digest, pp. 373-376, June
2003. [0014] Non Patent Literature 3: J. F. Buckwalter et al., "An
Injected Subharmonic Coupled-Oscillator Scheme for a 60-GHz
Phased-Array Transmitter", IEEE Transactions on Microwave Theory
and Techniques, Vol. 12, pp. 4271-4280, December 2006. [0015] Non
Patent Literature 4: S. Alausi et al., "A 60 GHz Phased Array in
CMOS", IEEE 2006 Custom Integrated Circuits Conference, Digest, pp.
393-396, San Jose, September 2006.
SUMMARY OF INVENTION
Technical Problem
[0016] In indoor millimeter-wave communication systems, the
following problem arises when the direct wave is shielded and the
radio transmission is to be continued by using a reflected
wave.
[0017] When the wave to be used (direct wave, reflected wave) is
switched, it is desirable to minimize the time during which the
transmission is disconnected. The reduction of the time during
which the transmission is disconnected becomes an important
requirement especially in the transmission of non-compressed
images, for example, in which the real-time capability is
indispensable. Meanwhile, in order to perform communication by
using a reflected wave, it is necessary to increase the directive
gain of the antenna, and thereby to increase the received signal
strength by narrowing the width of the antenna beam.
[0018] However, the number of directions (the number of steps) in
which the search needs to be performed increases as the beam width
becomes narrower. Therefore, the time necessary to find and set the
beam direction with which the incoming wave is effectively received
becomes longer, and therefore transmission-disconnected time also
becomes longer. Note that even in the case of apparatuses capable
of temporally storing received data, it is still undesirable in
practice because it requires a huge buffer memory to cope with a
long transmission-disconnected time. Accordingly, it has been
desired to develop a beam direction setting method that can shorten
the transmission-disconnected time when the direct wave is shielded
and the radio transmission is to be continued by using a reflected
wave.
[0019] FIG. 4 shows a configuration of an apparatus used in a beam
forming. Note that circuits that are inessential for the
explanation of the operation are omitted in the figure. A
transmitter 401 has a transmitting antenna array including m
antenna elements 405-1 to 405-m. A receiver 402 has a receiving
antenna array including n antenna elements 411-1 to 411-n. A
transmitting circuit 403 included in the transmitter 401 receives
transmission data from the outside of the circuit. The output of
the transmitting circuit 403 is branched into m signals, and they
are input to the respective amplitude/phase variable circuits 404-1
to 404-m. The respective signals input to the amplitude/phase
variable circuits 404-1 to 404-m are changed in their phases, and
eventually output from the transmitting antenna array composed of
the antenna elements 405-1 to 405-m. Furthermore, a
processing/arithmetic circuit 406 provides instructions on the
phase combination of the amplitude/phase variable circuits 404-1 to
404-m through a control circuit 407. With the phase change given to
each signal, it is possible to control the direction, the width,
and the like of the beam emitted from the transmitter 401.
Meanwhile, the receiver 402 has the reversed configuration to that
of the transmitter 401. That is, signals received by the receiving
antenna array composed of the antenna elements 411-1 to 411-n are
adjusted in their phases in amplitude/phase variable circuits 410-1
to 410-n, and then combined. A receiving circuit 409 demodulates
the combined signal, and externally outputs the received data.
Similarly to the processing/arithmetic circuit 406 in the
transmitter 401, a processing/arithmetic circuit 412 controls the
amplitude/phase given to each signal in the amplitude/phase
variable circuits 410-1 to 410-n. Note that the amplitude/phase
variable circuit is used to control the amplitude and the phase of
a signal that passes through the circuit.
[0020] FIG. 5 is a conceptual diagram for illustrating signal
states of the transmitter 401 and the receiver 402 shown in FIG. 4.
The transmitter 401 and the receiver 402 are linked through a MIMO
(Multi-Input Multi-Output) channel response matrix. It has been
known that if this channel response matrix is obtained, the optimal
setting of the amplitude and phase of a signal to be applied to the
antenna array of the transmitter-and-receiver (hereinafter called
"antenna weight vector") can be obtained by using SVD
(Singular-value Decomposition). However, on the other hand, since
SVD is complex and requires a long processing time, for example, it
is practically impossible to implement SVD for a non-compressed
image transmission apparatus in which the real-time capability is
indispensable.
[0021] In order to reduce the time necessary to determine the
antenna weight vector applied to the transmitter and the receiver,
Patent literature 3, for example, discloses a method for obtaining
the optimal phase at which the signal strength is maximized by
adding a unitary matrix (e.g., Hadamard matrix) as phases of the
antenna array and repeating the training of the antenna array of
the transmitter and the training of the antenna array of the
receiver. Although this method can reduce the necessary time in
comparison to SVD, it still requires a long time to obtain the
optimal antenna weight vector because the method repeatedly
performs the switching between the transmission and reception.
[0022] In particular, in a case where a link needs to be
re-established when disconnection of the transmission occurs in the
previously-established link, it is necessary to find another
optimal antenna weight vector in a shorter time in comparison to
the initial link establishment. Furthermore, in the case of
multipoint communication, it is also necessary to find the optimal
antenna weight vector in a short time because it requires the
re-establishment of multiple links.
[0023] The present invention has been made in view of the
above-described problems, and an object thereof is to provide a
radio control method capable of, when wireless communication is
implemented by performing beam forming, reducing the time necessary
to find and set a beam direction and thereby reducing the time
during which the transmission is disconnected.
Solution to Problem
[0024] A method according to a first exemplary aspect of the
present invention is a method of controlling a wireless
communication system which includes a transmitter having a
transmitting antenna array and a receiver having a receiving
antenna array. The control method includes the following processes
(a) to (d) that are performed when amplitudes and phases of signals
to be transmitted from at least two antenna elements among a
plurality of antenna elements constituting the transmitting antenna
array are independently controlled and amplitudes and phases of
signals to be received at least two antenna elements among a
plurality of antenna elements constituting the receiving antenna
array are independently controlled, the processes (a) to (d)
being:
(a) obtaining a channel response matrix by performing a training
process to obtain an optimal signal phase of the antenna array at
least between the transmitter and the receiver; (b) performing
singular-value decomposition process to decompose the channel
response matrix into a correlation matrix and eigenvalues; (c)
obtaining a diagonal matrix having square roots of the eigenvalues
obtained in the singular-value decomposition process as its
components; and (d) replacing all but one of diagonal components
included in the diagonal matrix with zeros, and obtaining an
antenna weight vector to be applied to the antenna array having
optimal communication quality for use in wireless communication
between the transmitter and the receiver based on a channel
response matrix reconstructed by using the component-replaced
diagonal matrix.
[0025] A wireless communication system according to second
exemplary aspect of the invention includes a transmitter and a
receiver. The transmitter includes a transmitting antenna array
having a plurality of transmitting antennal components, and is
configured to independently control amplitudes and phases of
transmission signals to be transmitted from at least two
transmitting antenna elements among the plurality of antenna
elements. Furthermore, the receiver includes a receiving antenna
array having a plurality of receiving antennal components, and is
configured to independently control amplitudes and phases of
received signals to be received at least two receiving antenna
elements among the plurality of antenna elements.
[0026] The transmitter and the receiver are configured so as to
perform an amplitude/phase control process of the transmitting and
receiving antenna arrays in cooperation. Note that the
amplitude/phase control process includes the following steps (a) to
(e):
(a) obtaining a channel response matrix by performing a training
process between the transmitter and the receiver; (b) performing
singular value decomposition to decompose the channel response
matrix into a correlation matrix and eigenvalues; (c) obtaining a
diagonal matrix having square roots of the eigenvalues obtained in
the singular-value decomposition as its components; (d) obtaining
an antenna weight vector to be applied to the antenna array having
optimal communication quality for use in wireless communication
between the transmitter and the receiver based on a channel
response matrix reconstructed by using a component-replaced
diagonal matrix, the component-replaced diagonal matrix being
obtained by replacing all but one of diagonal components included
in the diagonal matrix with zeros; and (e) controlling amplitudes
and phases of the transmission signals and received signals in
accordance with the antenna weight vector.
[0027] A transmitting apparatus according to third exemplary aspect
of the invention includes a transmitting antenna array and a
control unit. The transmitting antenna array includes a plurality
of antenna elements. Further, the control unit changes a beam
direction of the transmitting antenna array by controlling
amplitudes and phases of signals to be transmitted from at least
two antenna elements among the plurality of transmitting antenna
elements.
[0028] Furthermore, the control unit adjusts the beam direction by
supplying one antenna weight vector selected from a plurality of
antenna weight vectors to the transmitting antenna array, and
performs control such that the antenna weight vector to be supplied
to the transmitting antenna array is switched to a different one of
the plurality of antenna weight vectors in response to
deterioration in communication quality with a receiving apparatus.
Note that each of the plurality of antenna weight vectors
corresponds to one of a plurality of eigenpaths of a radio
transmission path between the transmitting apparatus and the
receiving apparatus, the plurality of eigenpaths being obtained by
performing a singular-value decomposition of a channel response
matrix with regard to the radio transmission path.
[0029] A receiving apparatus according to fourth exemplary aspect
of the invention includes a receiving antenna array and a control
unit. The receiving antenna array includes a plurality of antenna
elements. Further, the control unit changes a beam direction of the
receiving antenna array by controlling amplitudes and phases of
signals to be received by at least two antenna elements among the
plurality of antenna elements.
[0030] Furthermore, the control unit adjusts the beam direction by
supplying one antenna weight vector selected from a plurality of
antenna weight vectors to the receiving antenna array, and performs
control such that a phase combination to be supplied to the
receiving antenna array is switched to a different one of the
plurality of phase combinations in response to deterioration in
communication quality with a transmitting apparatus. Note that each
of the plurality of antenna weight vectors corresponds to one of a
plurality of eigenpaths of a radio transmission path between the
transmitting apparatus and the receiving apparatus, the plurality
of eigenpaths being obtained by performing a singular-value
decomposition of a channel response matrix with regard to the radio
transmission path.
Advantageous Effects of Invention
[0031] In accordance with each of the above-described exemplary
aspects of the present invention, when wireless communication is
implemented by performing beam forming, it becomes possible to find
and set a beam direction having excellent communication quality in
a short time.
BRIEF DESCRIPTION OF DRAWINGS
[0032] FIG. 1 shows transitions in a radio control procedure in
accordance with a first exemplary embodiment of the present
invention;
[0033] FIG. 2 shows transitions in a radio control procedure in
accordance with a second exemplary embodiment of the present
invention;
[0034] FIG. 3 shows transitions in a radio control procedure in
accordance with a third exemplary embodiment of the present
invention;
[0035] FIG. 4 shows an example of a device configuration used in
beam forming to which the present invention can be applied;
[0036] FIG. 5 is a schematic diagram for illustrating states of a
radio signal between a transmitter and a receiver;
[0037] FIG. 6 is a sequence diagram illustrating operations of a
transmitter and a receiver that are performed before actual
wireless communication in a radio control procedure in accordance
with a first exemplary embodiment of the present invention;
[0038] FIG. 7 is a sequence diagram illustrating operations of a
transmitter and a receiver that are performed when the wireless
communication is shielded in a radio control procedure in
accordance with a first exemplary embodiment of the present
invention;
[0039] FIG. 8A shows a figure for illustrating an aspect of
radio-wave propagation in a case where propagation paths are
created by local reflections of the radio signal in a radio control
procedure in accordance with a first exemplary embodiment of the
present invention (when no shielding occurs);
[0040] FIG. 8B shows a figure for illustrating an aspect of
radio-wave propagation in a case where propagation paths are
created by local reflections of the radio signal in a radio control
procedure in accordance with a first exemplary embodiment of the
present invention (when shielding occurs by a human body);
[0041] FIG. 9 shows a configuration of a system using wide-angel
antennas; and
[0042] FIG. 10 shows an example of a delay profile of a system
using wide-angle antennas when the system is used indoors.
DESCRIPTION OF EMBODIMENTS
First Exemplary Embodiment
[0043] A first exemplary embodiment of the present invention is
explained with reference to a transition diagram shown in FIG. 1.
Note that the wireless communication system in accordance with this
exemplary embodiment may employ a similar configuration to that
shown in FIG. 4. In S12, a transmitter 401 and a receiver 402
perform an initial training in order to optimize amplitude/phase
variable circuits 404-1 to 404-m and 410-1 to 410-n provided in the
transmitter 401 and receiver 402. In S13, a processing/arithmetic
circuit 406 or 412, or both of them in cooperation calculate a
plurality of candidate antenna weight vectors. The calculation
method for the plurality of candidate antenna weight vectors in S13
is described later. The obtained plurality of candidate antenna
weight vectors are recorded as a data string in storage circuits
408 and 414.
[0044] In S14, one candidate is selected from the plurality of
candidate phase combination obtained in S13 to perform
communication. In this process, it is preferable to select a
candidate antenna weight vector that is expected to provide the
best communication quality. the receiver 402 and the transmitter
401 monitor the communication state while communicating. The
monitoring of the communication state by the receiver 402 may be
implemented by measuring the communication quality by the receiving
circuit 409 or the processing/arithmetic circuit 412. For example,
communication quality such as a received-signal level, an SNR
(Signal to Noise Ratio), a BER (Bit Error Rate), a PER (Packet
Error Rate), and a FER (Frame Error Rate) may be measured.
Meanwhile, the monitoring of the communication state by the
transmitter 401 may be implemented by measuring the reception state
of a communication quality deterioration alert or the reception
state of a reception acknowledgement response (ACK) transmitted
from the receiver 402. Note that since publicly-known common
techniques may be employed for the actual technique for monitoring
the communication state, detailed explanation of the monitoring
technique in this exemplary embodiment is omitted.
[0045] When deterioration in the communication quality such as
disconnected communication is detected while the communication is
continued, the transmitter 401 and receiver 402 select another
antenna weight vector from the data string recorded in the storage
circuit 408 or 414 (S15).
[0046] In S16, it is determined whether the quality of the
communication using the newly selected antenna weight vector is
satisfactory or not. For example, the pass/fail of the
communication quality may be determined by measuring a
received-signal level, an SNR, or the like in the receiving circuit
409 or the processing/arithmetic circuit 412 included in the
receiver 402. When the communication quality is determined to be
satisfactory in S16, the transmitter 401 and receiver 402 return to
the communication state (S12). On the other hand, when the
communication quality is determined to be unsatisfactory in S16,
the transmitter 401 and receiver 402 transit to S16 to select
another antenna weight vector again.
[0047] When no antenna weight vector with which a satisfactory
communication state is achieved is found from the antenna weight
vectors recorded in the storage circuits 408 and 414, the process
returns to the initial training (S12) and is repeated from
there.
[0048] Next, a calculation procedure for a plurality of candidate
antenna weight vectors in S13 of FIG. 1 is explained hereinafter.
For the calculation of candidate antenna weight vectors, a MIMO
channel response matrix A is obtained by using a result of the
initial training in S12. The channel response matrix is expressed
by the following Formula (1).
A = [ a 11 a 12 a 1 m a 21 a 22 a 2 m a n 1 a n 2 a nm ] ( 1 )
##EQU00001##
[0049] A component A.sub.ij of the channel response matrix A
represents the response of a signal that is transmitted from ith
antenna 405-i of the transmitter 401 and received by jth antenna
411-j of the receiver 402. Further, the channel response matrix A
is m.times.n matrix where m is the number of antenna elements
included in the transmitter antenna array and n is the number of
antenna elements included in the receiver antenna array. The
channel response matrix A may be obtained, for example, by using
the method disclosed in Patent literature 3. Alternatively, the
channel response matrix A may be obtained by applying columns of a
unitary matrix in succession to an antenna weight vector while
transmitting a signal for training.
[0050] In this exemplary embodiment, a transmission signal vector T
and a received signal vector R are expressed by the following
Formulas (2) and (3).
T = [ t 1 t 2 t m ] ( 2 ) R = [ r 1 r 2 r n ] ( 3 )
##EQU00002##
In Formulas, a component t.sub.i of the transmission signal vector
T represents the input signal of ith amplitude/phase variable
circuit 404-i. Further, a component r.sub.i of the received signal
vector R represents the output signal of ith amplitude/phase
variable circuit 410-i. In the configuration example shown in FIGS.
4 and 5, since the signal from the transmitting circuit 403 is
equally branched into the amplitude/phase variable circuits 404-1
to 404-m, relations "t.sub.1=t.sub.2= . . . t.sub.m" and
"r.sub.1=r.sub.2= . . . r.sub.n" are satisfied.
[0051] Further, an antenna weight vector w.sub.t that is set to the
amplitude/phase variable circuits 404-1 to 404-m of the transmitter
401 is expressed by the following Formula (4). Furthermore, an
antenna weight vector w.sub.r that is set to the amplitude/phase
variable circuits 410-1 to 410-n of the receiver 402 is expressed
by the following Formula (5).
w t = [ .alpha. t 1 .theta. t 1 .alpha. 12 .theta. t 2 .alpha. tm
.theta. tm ] ( 4 ) w r = [ .alpha. r 1 - .theta. r 1 .alpha. r 2 -
.theta. r 2 .alpha. rn - .theta. en ] ( 5 ) ##EQU00003##
[0052] By using the definitions of the above-shown Formulas (1) to
(5), the signal response of the transmission/reception including
the amplitude/phase variable circuits 404-1 to 404-m on the
transmission side and the amplitude/phase variable circuits 410-1
to 410-n on the reception side is expressed by Formula (6) shown
below. In Formula (6), the matrix W.sub.t is a diagonal matrix that
has components of the antenna weight vector w.sub.t on the
transmission side as diagonal components. Further, the matrix
W.sub.r.sup.-1 in Formula (6) is the inverse matrix of a diagonal
matrix W.sub.r that has components of the antenna weight vector
w.sub.r on the reception side as diagonal components. The
definitions of the diagonal matrixes W.sub.t and W.sub.r are shown
in the following Formulas (7) and (8).
R = W r - 1 AW t T ( 6 ) W t .ident. [ .alpha. t 1 .theta. t 1 0 0
0 .alpha. t 2 .theta. t 2 0 0 0 .alpha. tm .theta. tm ] ( 7 ) W r
.ident. [ .alpha. r 1 - .theta. r 1 0 0 0 .alpha. r 2 - .theta. r 2
0 0 0 .alpha. rn - .theta. rn ] ( 8 ) ##EQU00004##
[0053] The channel response matrix A can be obtained by performing
training while changing the antenna weight vectors w.sub.t and
w.sub.r. Here, the definition of a correlation matrix is shown in
the following Formulas (9) and (10). Note that the index H of the
matrix indicates Hermitian transpose.
A.sup.HA (9)
AA.sup.H (10)
[0054] Letting .lamda..sub.1, .lamda..sub.2, . . . , .lamda..sub.M0
stand for eigenvalues of the correlation matrix in Formulas (9) and
(10) and letting .epsilon..sub.t,i and .epsilon..sub.r,i stand for
eigenvectors, the channel response matrix A can be decomposed as
shown in the following Formula (13). The decomposition process of
Formula (13) is called "singular value decomposition (SVD)".
E t = t , 1 , t , 2 , , t , M 0 ( 11 ) E r = r , 1 , r , 2 , , r ,
M 0 ( 12 ) A = E r DE t H = i = 1 M 0 .lamda. i r , i t , i H ( 13
) D = [ .lamda. 1 0 0 0 .lamda. 2 0 0 0 .lamda. M 0 ] ( 14 )
##EQU00005##
[0055] Note that M0 indicates the smaller one of the number m of
the transmission antenna elements and the number n of the reception
antenna elements. The eigenvector .epsilon..sub.t,i is an
eigenvector belonging to the eigenvalue .lamda..sub.i of the
symmetric matrix A.sup.HA shown in Formula (9), i.e., m.times.m
Hermitian matrix. E.sub.t is an eigenvector matrix having M0
eigenvectors E.sub.t,i as its components, and is expressed by
Formula (11). The eigenvector E.sub.r,i is an eigenvector belonging
to the eigenvalue .lamda..sub.i of the symmetric matrix AA.sup.H
shown in Formula (10), i.e., n.times.n Hermitian matrix. E.sub.r is
an eigenvector matrix having M0 eigenvectors .epsilon..sub.r,i as
its components, and is expressed by Formula (12). Further, D is a
diagonal matrix having the square roots of the eigenvalues
.lamda..sub.1, .lamda..sub.2, . . . , .lamda..sub.M0 as diagonal
components as shown in Formula (14). Note that each of the square
roots of the eigenvalues .lamda..sub.1, .lamda..sub.2,
.lamda..sub.M0 represents the energy of a respective one of M0
eigenpaths. The M0 eigenpaths have no correlation among them.
[0056] In this exemplary embodiment in accordance with the present
invention, all the diagonal components except for one component of
the matrix D in Formula (14) are replaced by zeros. Further, the
channel response matrix A is reconstructed by using the matrix
D.sub.i in which the components were replaced. For example, when
all the diagonal components expect for the second diagonal
component are replaced by zeros, the diagonal matrix D.sub.2 is
expressed by the following Formula (15). Further, the channel
response matrix A.sub.2 reconstructed by using the matrix D.sub.2
is expressed by Formula (16). From this channel response matrix
A.sub.2, one candidate antenna weight vector can be obtained.
D 2 = [ 0 0 0 0 .lamda. 2 0 0 0 0 ] ( 15 ) A 2 = E r D 2 E t H =
.lamda. 2 r , 2 t , 2 H ( 16 ) ##EQU00006##
[0057] By repeating the above-described procedure, M0 candidate
antenna weight vectors, at the maximum, corresponding to the
eigenvalues .lamda..sub.1, .lamda..sub.2, . . . , .lamda..sub.M0
respectively are obtained. The transmitter 401 and receiver 402
store at least a part of these M0 candidate antenna weight vectors
as a data string (database) in the storage circuits 408 and 414. As
described previously, the transmitter 401 and receiver 402 select
one antenna weight vector from the data string to start
communication (S13 and S14 in FIG. 1). Then, when the communication
using the optimal antenna weight vector selected in the early stage
deteriorates, the transmitter 401 and receiver 402 select the next
candidate from this data string (S15 in FIG. 1), validates the
communication quality (S16 in FIG. 1), and when the communication
quality is satisfactory, adopts that candidate (change from S13 to
S14).
[0058] Note that the square roots of the eigenvalues .lamda..sub.1,
.lamda..sub.2, . . . , .lamda..sub.M0 represent the energy of the
eigenpaths. Therefore, when the communication starts for the first
time in S14 of FIG. 1, it is preferable to select a candidate
antenna weight vector corresponding to the largest eigenvalue.
[0059] Further, all the diagonal component except for one component
of the matrix D shown in Formula (14) are replaced by zeros in the
above explanation of the calculation procedure of candidate antenna
weight vectors. However, a purpose of this exemplary embodiment is
to select one eigenpath and remove the influence by the other
eigenpaths. Therefore, it is not indispensable to replace the
diagonal components of the matrix D by zeros in the actual
calculation process. That is, other mathematically equivalent
methods may be used to achieve this purpose.
[0060] Next, operations of the transmitter 401 and receiver 402
performed in the state transition process shown FIG. 1 are
explained hereinafter in detail. FIG. 6 is a sequence diagram
illustrating operations of the transmitter 401 and receiver 402
performed in the transition process from S11 to S13 of FIG. 1,
i.e., in the process from the execution of the initial training to
the start of communication. Note that although not illustrated in
FIG. 4, it is preferable that the transmitter 401 and receiver 402
operate in synchronization with each other and that a transmission
path used to transfer information from the receiver 402 to the
transmitter 401 is provided. This transmission path in the reverse
direction may be a wireless transmission path or a wired
transmission path. Further, in the normal communication, the
transmitter 401 sends externally-input data to the receiver 402. On
the other hand, in the training, the processing/arithmetic circuit
406 makes the transmitting circuit 403 send a signal for training
(hereinafter called "training signal"). As a result, a training
signal is transmitted from the transmitter 401 to the receiver 402
in the training.
[0061] Hereinafter, each step of the sequence diagram of FIG. 6 is
explained one by one. Firstly, the transmitter 401 sets a phase for
training of the transmitter 401 in the amplitude/phase variable
circuits 404-1 to 404-m (S602-T), and sends a training signal
(S603-T). The transmitter 401 repeats the training signal sending
process, while changing the setting of the amplitude and phase for
the amplitude/phase variable circuits 404-1 to 404-m, until the
signal sending processes in all of the predetermined
amplitude-and-phase settings are completed (S604-T). During this
process, the receiver 402 receives the training signal (5603-R). In
the receiver 402, which has received the training signal, the
receiving circuit 409 measures the received-signal strength and/or
the received-signal quality and supplies data indicating the
measurement result to the processing/arithmetic circuit 412. The
processing/arithmetic circuit 412 processes the data indicating the
measurement result.
[0062] Next, the transmitter 401 sends a training signal for
training the receiver 402 (S606-T). During this process, the
receiver 402 sets a phase for training in the amplitude/phase
control circuits 410-1 to 410-n (S605-R), and receives the training
signal (S606-R). The receiver 402 repeats the training signal
receiving process until the signal receiving processes in all of
the predetermined amplitude-and-phase settings are completed
(S607-R).
[0063] In the step S608-R, the processing/arithmetic circuit 412
performs an SVD process by using measurement data obtained in the
steps S603-R to S607-R. Further, the processing/arithmetic circuit
412 obtains a plurality of antenna weight vectors (setting of the
amplitude and phase of a signal to be applied to the antenna array)
in accordance with the previously-described procedure, creates a
data string (database) including these candidate antenna weight
vectors, and stores the data string to the storage circuit 414
(S609-R). Further, the processing/arithmetic circuit 412 transmits
the created database to the transmitter 401 by using the
transmission path in the reverse direction (not shown) (S610-T, R).
The transmitter 401 stores the received database to the storage
circuit 408. At this point, the common contents are stored in both
the storage circuits 408 and 414. The transmitter 401 and receiver
402 select an optimal antenna weight vector from the common
databases, for example, in the descending order of the eigenvalues
(S611-T, R), set an amplitude and phase corresponding to the
selected antenna weight vector in the amplitude/phase variable
circuits, and start communication (S612-T, R).
[0064] Next, an operation performed when deterioration in the
communication quality such as disconnected communication occurs is
explained with reference to FIG. 7. FIG. 7 is a sequence diagram
illustrating operations of the transmitter 401 and receiver 402 in
the transition process from S14 to S16 of FIG. 1.
[0065] When a trouble such as disconnected communication occurs,
the receiver 402 detects deterioration in the communication quality
(S702-R), and notifies it to the transmitter 401 (S703-R). The
transmitter 401 receives the notification of the communication
quality deterioration from the receiver 402. Alternatively, the
transmitter 401 recognizes disconnected communication (or
deteriorated communication state) base on the fact that an ACK
signal, which would be transmitted from the receiver 402 upon
successful reception of data under normal communication
circumstances, is not received. In this state, each of the
transmitter 401 and receiver 402 obtains the next candidate antenna
weight vector from their own common databases (S704-T, R).
[0066] In a step S705-T, the transmitter 401 sets the next
candidate antenna weight vector in the amplitude/phase control
circuits 404-1 to 404-m. Similarly, in a step S705-T, the receiver
402 sets the next candidate antenna weight vector in the
amplitude/phase control circuits 410-1 to 410-n. After that, the
transmitter 401 and receiver 402 resume the communication (S706-T,
R). After the communication is resumed, the receiver 402 verifies
the communication quality (S707-R). When the communication quality
is satisfactory, the communication is continued, whereas when it is
unsatisfactory, the receiver 402 sends a notification indicating a
combination change (S708-R). The transmitter 401 continues the
communication unless the transmitter 401 receives a notification
indicating a combination change, or unless the transmitter 401
cannot receive an ACK signal from the receiver (S709-T). If not so,
the transmitter 401 and receiver 402 attempt to communicate by
using the next candidate antenna weight vector as long as there is
another candidate (S710-T, R). If the communication quality cannot
be improved with any of candidate antenna weight vector
combinations recorded in the storage circuits 408 and 414 and hence
there is no available candidate, the transmitter 401 and receiver
402 return to the initial training.
[0067] Incidentally, although the training in the transmitter 401
precedes the training in the receiver 402 in the exemplary
embodiment shown in FIG. 6, the training in the receiver 402 may be
performed before the training in the transmitter 401. Further,
although the SVD calculation (S608-R) and the database creation
(S609-R) are performed in the receiver 402 in the exemplary
embodiment in FIG. 6, at least one of them may be performed in the
transmitter 401. Specifically, received data or SVD result data may
be transferred from the receiver 402 to the transmitter 401, and
the processing/arithmetic circuit 406 may perform calculation using
these data. Further, as for the database creation, other cases
where antenna weight vectors obtained by a method other than the
method described in this specification are added in the database
also fall within the scope of this exemplary embodiment.
[0068] In accordance with this exemplary embodiment, when
deterioration in the communication quality such as disconnected
wireless communication or the like occurs, communication can be
swiftly resumed by selecting another candidate antenna weight
vector that is generated in advance. In other words, since it is
unnecessary to perform training and SVD calculation again every
time deterioration in the communication quality occurs in this
exemplary embodiment, it is possible to determine a new beam in a
very short time. Note that, in general, the SVD requires a large
quantity of calculation, and therefore even in this exemplary
embodiment, it is necessary to perform SVD calculation in the
initial training in order to establish a link. However, a longer
processing time is acceptable for the initial training in
comparison to the situation where the communication is recovered
after the occurrence of disconnected communication, and thus
causing little or no problem.
[0069] The following is supplementary explanation for the reason
why this method is effective for millimeter waves used indoors, or
microwaves having a frequency around or higher than 10 GHz and thus
having a high rectilinear propagation property. The propagation
paths that can be used for wireless communication are limited. That
is, only the direct wave and reflected waves from certain objects
such as walls, windows, and furniture can be used. Therefore,
angles at which waves should be emitted or angles at which waves
should be received are different from one wave to another.
Meanwhile, when subcarriers having a low rectilinear propagation
property such as 2.4 GHz micro waveband are used, it is necessary
to give consideration to effects caused by multiple scattering and
diffraction, and thus, in general, directional antennas are not
used. Therefore, situations are different between communication
using microwaves and millimeter waves having a frequency around or
higher than 10 GHz and communication using microwaves in the order
of 2.4 GHz. Note that there are some examples of development of
adaptive antennas having directivity for the purpose of removing
interference even in the field of communication using 2.4 GHz
microwaves. However, even if an adaptive-type directional antenna
is used, it is relatively easy to ensure satisfactory communication
quality at the angle of the direct wave or angles close to it in
the 2.4 GHz band because diffraction effects can be expected.
[0070] As for the indoor communication using beam forming in the
millimeter waveband, it is necessary to take the following
characteristics into consideration. As described previously, the
number of reflected waves other than the direct wave is limited.
Further, even if a certain direct wave or a reflected wave is
shielded by an obstacle (e.g., human body), there is no correlation
between the shielded wave and other waves. Therefore, in the
millimeter wave communication system, as described in this
exemplary embodiment, it is possible to obtain a reserve beam
direction while performing communication in a beam direction having
the best communication state. In contrast to this, when the
frequency is lower than around 10 GHz, contribution of multiple
reflections and diffractions to the communication quality is large.
Therefore, even if a directional antenna is used, the propagation
state of the reserve beam direction varies depending on the
presence/absence of an obstacle. That is, there is a high
possibility that the received signal state of the reserve beam
direction, which has satisfactory quality without any obstacle, is
changed due to the presence of an obstacle. Therefore, it is
difficult to obtain the advantageous effect of the present
invention in 2.4 GHz microwave communication and the like.
[0071] Further, in millimeter wave communication, a propagation
path may be sometimes created by local reflection. FIGS. 8A and 8B
show aspects of such a situation. In FIG. 8A, there are a
transceiver 81 and a receiver 82, and it is assumed that there are
propagation paths in the beam forming including a direct wave A, a
local reflected wave B, and a reflected wave C propagating through
a long path. As shown in FIG. 8B, there is a possibility that the
direct wave A and the local reflected wave B are simultaneously
shielded, for example, by a human body. When there is a high
correlation between the propagation path A (direct wave A) and the
propagation path B (reflected wave B), they are not decomposed by
SVD. Therefore, the same antenna weight vector is applied to the
propagation paths A and B. Therefore, it is possible to eliminate
candidate antenna weight vectors that are simultaneously shielded
in this exemplary embodiment. However, when the correlation between
the propagation paths A and B is low, they could become different
candidate antenna weight vectors. However, even in such a case, the
only necessary process to be added is to perform one extra
repetition of the steps (S704-T, R) to (S710-T, R). Therefore, the
time required for the recovery of the communication can be still
significantly reduced in comparison to the case where the training
itself is performed again.
Second Exemplary Embodiment
[0072] A second exemplary embodiment of the present invention is
explained with reference to a transition diagram shown in FIG. 2.
Note that the wireless communication system in accordance with this
exemplary embodiment may employ a similar configuration to that
shown in FIG. 4. Each state from S21 to S26 shown in FIG. 2 as well
as their transition conditions are similar to those described in
the first exemplary embodiment and shown as S11 to S16 in FIG. 1.
Therefore, detailed explanations of S21 to S26 are omitted.
[0073] In S27 of FIG. 2, the processing state is changed from the
communication continuation state (S24) to perform an additional
second training. The second training may be performed at regular
intervals, or may be performed during idle times in which no data
is transmitted/received.
[0074] In S28, the processing/arithmetic circuit 406 or 412, or
both of them in cooperation re-calculate a plurality of candidate
antenna weight vectors. The processing/arithmetic circuits 406 and
412 update the data string stored in the storage circuits 408 and
414 with the plurality of candidate antenna weight vectors obtained
by the recalculation.
[0075] In this exemplary embodiment, the plurality of candidate
antenna weight vectors are updated by examining the situation in
regard to the reserve beam direction by performing the second
training at regular intervals or as necessary. In this way, the
wireless communication system in accordance with this exemplary
embodiment can obtain the latest candidate antenna weight vectors
at all times. Note that the second training (S27) may be divided
into multiple sections so that they can be performed in intervals
of the communication. In this way, there is no need to suspend the
communication for a long time. Further, when the communication is
disconnected or the communication quality deteriorates, it is
desirable to restore the communication in a very short time.
However, since so much immediacy is not required for the second
training, SVD and the like can be performed without causing any
substantial problem.
Third Exemplary Embodiment
[0076] A third exemplary embodiment of the present invention is
explained with reference to a transition diagram shown in FIG. 3.
The wireless communication system in accordance with this exemplary
embodiment may employ a similar configuration to that shown in FIG.
4. Further, the wireless communication system in accordance with
third exemplary embodiment performs the same operations as those of
the second exemplary embodiment. That is, each state from S31 to
S38 shown in FIG. 3 as well as their transition conditions are
similar to those described in the second exemplary embodiment and
shown as S21 to S28 in FIG. 2. Therefore, detailed explanations of
S31 to S38 are omitted.
[0077] In this exemplary embodiment, when deterioration in the
communication quality such as disconnected communication or the
like occurs, the next candidate antenna weight vector is selected
from the plurality of candidates recorded in the database (S35) and
a fine adjustment is performed in that state (S39). This fine
adjustment means a method for searching for an optimal beam without
taking much time. Specifically, the adjustment may be performed by
slightly changing the beam or the set phase so that better
communication quality is obtained. Furthermore, simplified beam
searching procedure such as "Beam Tracking" disclosed in Patent
literature 3 may be applied.
[0078] For example, in the case where the candidate antenna weight
vector is changed in order from an antenna weight vector
corresponding to a large eigenvalue to an antenna weight vector
corresponding to a small eigenvalue as described in detail in the
first exemplary embodiment, the received power gradually decreases
and the accuracy could gradually deteriorate. Accordingly, it can
provide such an advantageous effect that an antenna weight vector
with which stable transmission can be achieved with high accuracy
can be found by performing the fine adjustment, e.g. adjusting a
receiving antenna gain, where the received power is weakened by the
occurrence of shielding.
Fourth Exemplary Embodiment
[0079] A fourth exemplary embodiment is characterized in that the
training and the acquisition/setting of antenna weight vectors are
performed at a low rate (with a narrow band) and actual
communication is performed at a relatively high rate (with a wide
band). For the other operations, the method described in one of
first to third exemplary embodiments may be employed.
[0080] Since the free space propagation loss is large in millimeter
wave communication, the received power is expected to be small.
Therefore, if a unitary matrix is set as antenna weight vectors of
the amplitude/phase variable circuits 404-1 to 404-m or 410-1 to
410-n, there is a possibility that a sufficient CNR (Carrier to
Noise Ratio) is not achieved. Accordingly, it is expected that the
use of the low rate (narrow band) having better reception
sensitivity provides advantageous effects such as enabling the
training and improving the accuracy. Note that the use of low rate
(narrow band) means to narrow the frequency band used to transmit a
training signal in order to narrow the noise bandwidth, or to adopt
a modulation technique having a small necessary CNR. Note also that
"to adopt a modulation technique having a small necessary CNR"
means, in other words, to adopt a modulation technique having a
large distance between signal points on the constellation
(typically a smaller transmission rate). Note also that it is
assumed that a narrow beam width is used in this exemplary
embodiment, and therefore there is no significant difference in
optimal beam (or corresponding antenna weight vector) regardless of
whether the transmission is a low rate (narrow band) or a high rate
(wide band) because the correlative bandwidth is wide.
[0081] In the above four exemplary embodiments, the term
"communication quality" is used. The communication quality may be,
for example, any parameter representing communication quality such
as a received-signal level, an SNR (Signal to Noise Ratio), a BER
(Bit Error Rate), a PER (Packet Error Rate), and a FER (Frame Error
Rate), and one or more than one of them may be used. Further, a
certain data string in a preamble included in a transmission data
string of the transmitter 401 may be used for the evaluation of
communication quality.
[0082] Further, the amplitude/phase variable circuits 404-1 to
404-m or 410-1 to 410-n are used in the above-described four
exemplary embodiments. However, the essential purpose is to realize
desired antenna weight vectors, and it can be constructed by using
any appropriate circuits.
[0083] Furthermore, although the transmitter 401 and receiver 402
are described as separate devices in the explanation of the
above-described four exemplary embodiments, needless to say, the
present invention can be also applied to communication between
transmission-and-reception devices each having a transmitting
function and a receiving function. In such a case, if the
single-piece transmission-and-reception antenna is used, the
training needs to be performed only in one of the two directions of
the radio transmission path because of the principle of
reciprocity.
[0084] Incidentally, control and arithmetic processing for the
generation/switching of a plurality of candidate antenna weight
vectors that are performed by the transmitter 401 in the
above-described first to fifth exemplary embodiments can be also
implemented by executing a computer program(s) for
transmitter/receiver control in a computer such as a
microprocessor. In the case of the first exemplary embodiment, for
example, processing in the steps S703-T to S705-T and S708-T to
S710-T shown in the flowchart of FIG. 7 may be performed in a
computer executing a transmitter control program. Similarly,
control and arithmetic processing for the generation/switching of a
plurality of candidate antenna weight vectors that are performed in
the receiver 402 can be also implemented by executing a computer
program(s) for transmission/reception control in a computer such as
a microprocessor. In the case of the first exemplary embodiment,
for example, processing in the steps S702-R to S705-R and S707-R to
S710-R shown in the flowchart of FIG. 7 may be performed in a
computer executing a receiver control program. These transmitter
control program and receiver control program can be stored in
various types of computer-accessible storage media. Further, these
programs can be also transmitted through communication media. Note
that examples of the storage media include a flexible disk, a hard
disk, a magnetic disk, magneto-optic disk, a CD-ROM, a DVD, a ROM
cartridge, a RAM memory cartridge with a battery backup, a flash
memory cartridge, and a nonvolatile RAM cartridge. Further,
examples of the communication media include a wire communication
medium such as a telephone line, a wireless communication medium
such as a microwave line, and the Internet.
[0085] Furthermore, in addition to the processing/arithmetic
circuits 406 and 412, a part of the transmitting circuit 403
(modulation processing and the like), a part of the receiving
circuit 409 (demodulation processing and the like), and other
components for digital signal processing or device control such as
the control circuit 407 and control circuit 413 may be implemented
by a computer such as a microcomputer and a DSP (Digital Signal
Processor). Furthermore, the so-called "software-antenna
technology" may be applied to the transmitter 401 and receiver 402.
Specifically, the amplitude/phase variable circuits 404-1 to 404-m
and 410-1 to 410-n may be constructed by a digital filter(s), or a
computer such as a DSP.
[0086] Further, the present invention is not limited to the
above-described exemplary embodiments, and needless to say, various
modifications can be made without departing from the
above-described spirit of the present invention.
[0087] This application is based upon and claims the benefit of
priority from Japanese patent application No. 2008-184405, filed on
Jul. 16, 2008, the disclosure of which is incorporated herein in
its entirety by reference.
REFERENCE SIGNS LIST
[0088] 401, 81, 91 TRANSMITTER [0089] 402, 82, 92 RECEIVER [0090]
403 TRANSMITTING CIRCUIT [0091] 404-1.about.m AMPLITUDE/PHASE
VARIABLE CIRCUIT [0092] 405-1.about.m TRANSMITTING ANTENNAL ARRAY
[0093] 406 PROCESSING/ARITHMETIC CIRCUIT [0094] 407 CONTROL CIRCUIT
[0095] 408 STORAGE CIRCUIT [0096] 409 RECEIVING CIRCUIT [0097]
410-1.about.n AMPLITUDE/PHASE VARIABLE CIRCUIT [0098] 411-1.about.n
RECEIVING ANTENNAL ARRAY [0099] 412 PROCESSING/ARITHMETIC CIRCUIT
[0100] 413 CONTROL CIRCUIT [0101] 414 STORAGE CIRCUIT [0102] 83
BEAM PATTERN (IMAGE) [0103] 84, 85 REFLECTOR [0104] 86 HUMAN
BODY
* * * * *